Open APIs Crucial

Olivier Travers writes of the need for open APIs. We at Identity Commons consider open APIs to be crucial (as well as open governance, open privacy and security mechanisms, etc.). Not only are the technologies used by 2idi to implement the IC platform based upon open standards and code (LAMP, XRI, XDI and SAML/Lasso), but all 2idi core software will be dual (BSD/GPL) licensed. In addition, we will be explicitly working with service providers of all types to create more and better open APIs so that they can more easily use the open authentication model. It may sound counterintuitive, but not only do we want to have as many people and applications connecting to us as possible, but we also want to encourage a multitude of providers offering services compatible with ours - all part of our open source plan for success.

As there will be other identity models in existence for a while (after all, we're coming in rather late to the game) we will work to create interoperability where possible. Of course, since 2idi/Identity Commons is the only initiative that I know of that is aimed at a fully open system giving its users total control over their identity - including where their information is stored - some of the other identity providers that we interface with may have less than the desired level of compatibility.

Comments

> 2idi/Identity Commons is

> 2idi/Identity Commons is the only initiative that I know of
> that is aimed at a fully open system giving its users total
> control over their identity -
> including where their information is stored

Shibboleth?
http://shibboleth.internet2.edu/

Shibboleth is the closest

Shibboleth is the closest thing to what we are building, and I really like the work that they are doing. In recent talks with Scott Cantor (one of Shib's architects) we agreed that the concept of a universally resolvable i-name could enhance the initial authentication process used by Shib. We're also in conversation with Internet2 about chaordic governance mechanisms for federated communities. Finally, like Shib, we're basing our authentication on SAML - though we're leaning towards the 2.0/Liberty ID-FF v1.2 spec.

Beyond this, we plan to provide lightweight i-broker client and data hosting services that anyone with a machine connected to the Internet could run, increasing the opportunity for people to become their own data "banks", Our still-in-committee standard for data access negotiation, XDI, may also prove to be an area that Shib follows our lead.

But bottom line, you are correct in noticing the close similarity between 2idi/Identity Commons and Shibboleth. My statement was perhaps a little overly broad.

Hi, Fen! Interesting project

Hi, Fen!

Interesting project you have going here. I have a mix of anticipation and skepticism, and i definitely want to understand it better. I'm sorry now that i didn't have the chance to discuss this with you at Hackers. I'm in Berkeley and have some free time over the next week; would you be willing to meet and talk?

(Note: I'd like to send this message to you directly, but i don't have your e-mail address on hand. So, i click your name to contact you, thinking "Aha! Now i get to try out this system for real." I get a nice form for explaining why i'm contacting you so you can vet me. But the system is closed! I'm not allowed to contact you by your i-name unless i get my own i-name first. Thus, it appears that i have to give $25 to your project just to contact you. Not that your project isn't worth supporting — it looks pretty good, in fact — but i submit that this may not be the kind of first impression you want to present.)

(I'll email you separately

(I'll email you separately regarding getting together.)

Point well taken regarding the requirement I had for an i-name - I have removed that and you can now contact me with a valid email address (which will be verified) if you don't have an i-name.

I had thought that it would be more "viral" to require i-names, but until free, community i-names are available (by next year - only a few more weeks to wait) I agree that it is anti-social to force people to pay in order to contact me. And I try to be social.

Thanks for the tip!